Multi-drug resistance due to efflux is one of the most pressing issues in treating bacterial infections. Antibiotics are extruded from the cell and cannot reach high enough intracellular concentrations to exert a therapeutic effect. While efforts to address this problem have focused on targeting one efflux pump at a time, resistance mutations can quickly develop. We propose to target TrmD to reduce efflux at multiple pumps simultaneously so as to accelerate bactericidal action. TrmD is a bacterial-specific S-adenosyl-methionine (AdoMet)-dependent tRNA methyl transferase that controls the accuracy of protein synthesis on the reading frame. Loss of TrmD leads to accumulation of +1 frameshift (+1FS) errors, which cause pre-mature termination of protein synthesis. We recently discovered that multiple efflux genes in E. coli and in other Gram (-) bacteria contain TrmD-dependent codons near the AUG start codon of the reading frame. We therefore hypothesize that targeting TrmD can inactivate protein expression of all of these genes. By reducing drug efflux of multiple pumps at once, we propose that targeting TrmD offers a novel solution to an unmet medical need. Successful targeting TrmD requires an understanding of its AdoMet domain and the quality control mechanism maintained by its product m1G37-tRNA. Using E. coli TrmD (EcTrmD) as a model, we provide these prerequisites in the following three aims. In Aim 1, we will develop a molecular-level understanding of the AdoMet domain of TrmD. This domain has the unusual ability to maintain the methylation activity even at low levels of AdoMet. We will test the hypothesis that this ability is conferred by the unusual topological protein knot-fold of the domain that bends the methyl donor upon binding. Using both kinetic and cellular assays, we will determine how the protein knot-fold uses AdoMet binding to facilitate tRNA binding and to promote methyl transfer. We hypothesize that this facile coordination between AdoMet binding and catalytic activity is at the root of the unique biology of the domain. In Aim 2, we will develo a mechanistic-level understanding of how the m1G37-tRNA product of TrmD improves the accuracy of protein synthesis on the ribosome. Accuracy will be determined by the ability of m1G37-tRNA to reduce +1FS errors at slippery mRNA sequences and to reduce decoding errors of cmo5U34, a wobble modification that frequently accompanies m1G37 in natural tRNAs. In Aim 3, we will develop a cellular-level understanding of how inactivation of TrmD reduces synthesis of efflux proteins. We hypothesize that this reduction of efflux proteins will increase intracellular accumulation of traditional antibiotics, leading to faster bactericidal actin. We will also define the scope of TrmD activity using ribosome profiling to identify genes whose protein expression is arrested in TrmD deficiency. This will provide the basis for new strategies of antibiotic targeting and new paradigms for antibiotic discovery.